Zoltar is my NFL football machine learning prediction system. It’s a hybrid system that uses a custom “reinforcement learning” algorithm plus a neural network. Here are Zoltar’s predictions for week #10 of the 2017 NFL season:
Zoltar: seahawks by 0 dog = cardinals Vegas: seahawks by 5.5 Zoltar: steelers by 4 dog = colts Vegas: steelers by 10 Zoltar: lions by 16 dog = browns Vegas: lions by 10 Zoltar: saints by 0 dog = bills Vegas: saints by 2.5 Zoltar: titans by 6 dog = bengals Vegas: titans by 5 Zoltar: vikings by 0 dog = redskins Vegas: vikings by 1.5 Zoltar: jaguars by 4 dog = chargers Vegas: jaguars by 4 Zoltar: packers by 2 dog = bears Vegas: bears by 3 Zoltar: buccaneers by 4 dog = jets Vegas: jets by 1.5 Zoltar: rams by 5 dog = texans Vegas: rams by 12 Zoltar: giants by 5 dog = fortyniners Vegas: giants by 1 Zoltar: cowboys by 0 dog = falcons Vegas: falcons by 3 Zoltar: patriots by 5 dog = broncos Vegas: patriots by 7.5 Zoltar: panthers by 2 dog = dolphins Vegas: panthers by 9
Zoltar theoretically suggests betting when the Vegas line is more than 3.0 points different from Zoltar’s prediction. Typically, Zoltar has three or four hypothetical pieces of recommendation, but for week #10, Zoltar has eight pieces of advice — probably due to the very large number of injuries in mid-season.
1. Zoltar likes the Vegas underdog Cardinals against the Seahawks. Vegas has the Seahawks as 5.5 point favorites but Zoltar thinks the game is evenly matched. Therefore, a bet on the Cardinals will pay off if the Cardinals win by any score, or if the Seahawks win but by less than 5.5 points (in other words, 5 points or less).
2. Zoltar likes the Vegas underdog Colts against the Steelers.
3. Zoltar likes the Vegas favorite Lions against the Browns.
4. Zoltar likes the Vegas underdog Packers against the Bears.
5. Zoltar likes the Vegas underdog Buccaneers against the Jets.
6. Zoltar likes the Vegas underdog Texans against the Rams.
7. Zoltar likes the Vegas favorite Giants over the 49ers.
8. Zoltar likes the Vegas underdog Dolphins against the Panthers.
There are significant injuries in five of these eight games. I have an advanced version of Zoltar that takes injuries into account, and Advanced Zoltar has only four suggestions.
Zoltar had a mediocre showing last week. Against the Vegas spread, which is what Zoltar is designed to predict, Zoltar went only 1-1 (one game, the Dolphins vs. Raiders was a push when the Vegas spread of 3.0 points exactly matched the actual 27-24 score).
For the 2017 season so far, against the Vegas point spread, Zoltar is a pretty good 24-11 (68% accuracy). If you must bet $110 to win $100 (typical in Vegas) then you must theoretically predict with 53% or better accuracy to make money, but realistically you must predict at 60% or better accuracy.
Just for fun, I also track how well Zoltar does when only predicting which team will win. This isn’t really useful except for parlay betting. For week #9, Zoltar was a poor 8-5 just predicting winners.
For comparison purposes, I also track how well Bing and the Vegas line do when just predicting who will win. In week #9, Bing was a good 10-3, and Vegas was also good at 9-3 (again, one push) when just predicting winners.
For the 2017 season so far, just predicting the winning team, Zoltar is 83-49 (63% accuracy), Bing is also 83-49 (63% accuracy), and Vegas is 79-49 (62% accuracy). The best humans are typically about 67% accurate predicting winners.